[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2432":3},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":23,"tags":25,"view_count":30,"doi":31,"paper":32,"created_at":52},2432,"Multi-annual sub-pixel land cover mapping with TESSERA latent embeddings","https:\u002F\u002Fdoi.org\u002F10.31223\u002Fx59j7c","Mapping land cover in highly heterogeneous landscapes is challenging, and classifications have inherent limitations where the spatial resolution of remotely sensed data exceeds the size of small objects. Sub-pixel land cover fraction maps based on medium-resolution optical remote sensing data like Landsat or Sentinel-2 overcome this limitation but are often not consistently available across multiple years because creating temporally robust and comparable multi-class fraction models can be challenging. Latent embeddings from geospatial foundation models are promising as they represent high-dimensional, large-scale general-purpose spectral, temporal, and structural features derived from Earth Observation data that are largely independent of available high-quality observations at specific points in time. We here assessed the usability of TESSERA latent embeddings for machine-learning regression-based land cover fraction models and compared the performance to more conventional spectral-temporal metrics and spline coefficients as input features. We also assessed the suitability of these inputs to train temporally transferred and generalized fraction models. We found that TESSERA latent embeddings are a well-suitable input to land cover fraction mapping, performing slightly better than spectral-temporal-metrics-based models in many cases, both being notably outperformed by spline coefficients (MAE across classes and years 11.58 vs. 11.64 vs. 9.45). Spline coefficients also performed best in all temporally generalized and many temporally transferred models. However, models trained with TESSERA latent embeddings demonstrate the highest consistency of transferred models, highlighting their capability to handle data gaps. We suggest that TESSERA latent embeddings are a valuable input to fraction mapping where data availability across years is highly variable, but spline coefficients generally perform best when sufficient high-quality observations are available.","在高度异质的景观中绘制土地覆盖图具有挑战性，而且当遥感数据的空间分辨率超过小对象尺寸时，分类存在固有局限。基于中分辨率光学遥感数据（如Landsat或Sentinel-2）的亚像元土地覆盖比例图克服了这一局限，但往往无法在多个年份间一致获取，因为构建时间上稳健且可比的多类比例模型可能颇具挑战。地理空间基础模型的潜在嵌入（latent embeddings）前景广阔，因为它们代表了从地球观测数据中提取的高维、大规模通用光谱、时间和结构特征，这些特征在很大程度上独立于特定时间点可获取的高质量观测。我们在此评估了TESSERA潜在嵌入用于基于机器学习回归的土地覆盖比例模型的可用性，并将其性能与更传统的光谱-时间指标（spectral-temporal metrics）和样条系数（spline coefficients）作为输入特征进行了比较。我们还评估了这些输入用于训练时间迁移和广义化比例模型的适用性。我们发现，TESSERA潜在嵌入是土地覆盖比例制图的合适输入，在许多情况下略优于基于光谱-时间指标的模型，而两者的表现均明显不及样条系数（跨类别和年份的MAE分别为11.58、11.64和9.45）。样条系数在所有时间广义化模型和许多时间迁移模型中也表现最佳。然而，使用TESSERA潜在嵌入训练的模型在迁移模型间表现出最高的一致性，凸显了其处理数据缺口的能​​力。我们建议，在跨年份数据可用性高度可变的情况下，TESSERA潜在嵌入是比例制图的宝贵输入，但当有足够的高质量观测可用时，样条系数通常表现最佳。",null,"OpenAlex","2026-09-11T00:00:00Z","论文",10,false,70,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":22},12,21,17,8,1,"评估TESSERA地理空间基础模型潜嵌入用于多年份亚像元土地覆盖制图，方法新颖、结论可靠，对农业遥感监测有参考价值，但属细分方法研究，影响面有限。",[24],{"name":10,"url":6},[26,27,28,29],"农业人工智能","遥感","基础模型","土地覆盖",0,"10.31223\u002Fx59j7c",{"doi":31,"openalex_id":33,"authors":34,"venue":9,"cited_by_count":30,"oa_url":44,"card":45,"direction":49,"ingested_from":51},"W7212309976",[35,38,40,42],{"name":36,"orcid":37},"Franz Schug","https:\u002F\u002Forcid.org\u002F0000-0003-1534-5610",{"name":39,"orcid":9},"David Klehr",{"name":41,"orcid":9},"Jari Mahler",{"name":43,"orcid":9},"David Frantz","https:\u002F\u002Feartharxiv.org\u002Frepository\u002Fobject\u002F14912\u002Fdownload\u002F25925\u002F",{"tldr":46,"method":47,"finding":48,"direction":49,"opportunity":50},"评估TESSERA潜嵌入用于多年度亚像元土地覆盖制图，并与光谱时序指标和样条系数比较。","使用TESSERA潜嵌入、光谱时序指标和样条系数作为特征，训练机器学习回归模型预","样条系数精度最高，但TESSERA潜嵌入在时间迁移模型中一致性最好，适合数据缺失场景。","农业遥感与作物表型","可探索将地理空间基础模型潜嵌入用于多年度作物覆盖比例制图，解决数据不均下的时序泛化问题。","openalex","2026-09-14T23:30:27.289645Z"]